Increased ghrelin signaling prolongs survival in mouse models of human aging through activation of sirtuin1

Caloric restriction (CR) is known to retard aging and delay functional decline as well as the onset of diseases in most organisms. Ghrelin is secreted from the stomach in response to CR and regulates energy metabolism. We hypothesized that in CR ghrelin has a role in protecting aging-related diseases. We examined the physiological mechanisms underlying the ghrelin system during the aging process in three mouse strains with different genetic and biochemical backgrounds as animal models of accelerated or normal human aging. The elevated plasma ghrelin concentration was observed in both klotho-deficient and senescence-accelerated mouse prone/8 (SAMP8) mice. Ghrelin treatment failed to stimulate appetite and prolong survival in klotho-deficient mice, suggesting the existence of ghrelin resistance in the process of aging. However, ghrelin antagonist hastened death and ghrelin signaling potentiators rikkunshito and atractylodin ameliorated several age-related diseases with decreased microglial activation in the brain and prolonged survival in klotho-deficient, SAMP8 and aged ICR mice. In vitro experiments, the elevated sirtuin1 (SIRT1) activity and protein expression through the cAMP–CREB pathway was observed after ghrelin and ghrelin potentiator treatment in ghrelin receptor 1a-expressing cells and human umbilical vein endothelial cells. Furthermore, rikkunshito increased hypothalamic SIRT1 activity and SIRT1 protein expression of the heart in the all three mouse models of aging. Pericarditis, myocardial calcification and atrophy of myocardial and muscle fiber were improved by treatment with rikkunshito. Ghrelin signaling may represent one of the mechanisms activated by CR, and potentiating ghrelin signaling may be useful to extend health and lifespan.

immunohistochemical and gene expression studies were collected at 19 weeks after treatment with rikkunshito in SAMP8 mice.
Experiment 3: Rikkunshito (1000 mg/kg, p.o.) was daily administered to 18-week-old SAMP8 mice for 4 days, and tissue samples were collected 2 hours after administration for SIRT1 analysis.

Animal experiments (ICR mice)
Experiment 1: In this experiment, 16-to 18-month-old ICR mice were given rikkunshito (0.5, 1%)-containing chow or control chow in individual houses. Because of a large difference in age, the aged group of ICR mice was retrospectively assessed using a grading score with accelerating aging (> 1.0) and body weight (> 50 g) at the start of the study. At 2 months after treatment with rikkunshito, a passive avoidance test, the elevated plus-maze test, and the open-field test were performed. Median survival was calculated using a Kaplan-Meier plot. After death in the survival study, heart samples were collected for the histochemical analysis.
Experiment 2: Fasting blood and tissue samples for SIRT1 analysis were collected in 26-month-old ICR mice treated with rikkunshito (1%) for 8 months or 4-month-old ICR mice.
Experiment 3: Brain samples for immunohistochemical study were collected 4 weeks after treatment with rikkunshito (1%) in 12-month-old ICR mice.

Animal experiments (GHS-R knockout mice)
Twelve-week-old GHS-R knockout mice, heterozygous mice, and wild type C57BL/6 mice were treated with rikkunshito (1%)-containing chow or control chow. After 4 weeks, hypothalamic samples in these mice were collected.

Food intake, body weight, and aging score
The rate of change in food intake, body weight, food efficiency (calculated as body weight gain per food intake every five weeks), and grading score with accelerating aging during the survival study were obtained from least squares analysis.

Locomotor activity
The locomotor activity of mice in the home cage was measured during a light-dark cycle with lights on from 7:00 to 19:00 with an infrared sensor (NS-AS01; Neuroscience, Inc., Tokyo, Japan).
Step-through passive-avoidance test The apparatus (Neuroscience, Inc.) for the step-through passive-avoidance test consisted of two compartments, one illuminated (100 mm x 120 mm x 145 mm) and the other dark (140 mm x 185 mm x 190 mm), which were separated by a guillotine door. A mouse was placed in the illuminated compartment and stepped through the open guillotine door into the dark compartment and was given a foot shock (0.3 mA) for three seconds. Such trials were performed once a day, and the time spent in the illuminated compartment was defined as the latency period.

Open-field test
The open-field test consisted of a square floor (50 cm × 50 cm) enclosed by walls 25 cm high and divided into 25 areas of 10 cm intervals. A mouse was placed in the center part of the open field, and the total number of line crossings in areas and total number of entries into the central part for 5 min were determined using the analysis software LimeLight (Neuroscience, Inc.).

Elevated plus-maze test
The elevated plus-maze test consisted of two open arms and two enclosed arms (20 cm × 5 cm each), arranged so that the arms of the same type were opposite each other and elevated to a height of 50 cm. A mouse was placed in the central square of the elevated plus-maze, and the number of entries into the open arms and the time spent in the open arms in the plus-maze in 5 min were measured using the analysis software LimeLight.

Histochemical study
After death or euthanasia at the end of the survival study, several tissue samples including heart and gastrocnemius muscle were fixed in 10% phosphate-buffered formalin, paraffin embedded, and stained with hematoxylin and eosin for light microscopic examination. Scores were obtained using a semi-quantitative pathological scoring system (None; 0, Minimal; +1, Mild; +2, Moderate; +3). For brain samples, serial sections of 10 μm thickness were mounted onto MAS-coated glass slides, dewaxed with xylene and processed through ethanols to water. The sections were subsequently incubated with anti-Iba-1 rabbit polyclonal antibody (Wako Pure Chemical Industries), and then processed according to the peroxidase-labeled antibody method. The products were visualized in a reaction with 3.3'-diamino-benzidine (DAB) and H 2 O 2 . Stained sections were observed equipped under a light microscope with a color-chilled 3CCD camera. The number of amoeboid microglia positively stained with anti-Iba1 antibody was quantitatively analyzed under a light microscope. The five 400 x 600 μm squares within the identical brain area (2 mm square) beneath the corpus callosum of mice were blindly captured, and the number of activated microglia with amoeboid morphology were counted and analyzed statistically (Prism 6, GraphPad Software Inc., La Jolla, CA).

Electrophysiologic study
The afferent activity of the gastric vagus nerve and the sympathetic nerve activity of the brown adipose tissues in urethane anesthetized rats were recorded via a pair of silver wire electrodes. A rate meter with a rest time of 5 s was used to observe the time course of nerve activity. Ghrelin (10 ng/rat) and rikkunshito (1000 mg/kg) or its constituents (400 mg/kg) were administered through a catheter inserted into the inferior vena cava and the duodenum, respectively. The mean numbers of impulses per 5 s over 50 s before and after the injection were compared.

Cell culture and transfection
293-GHS-R cells and 293-Mock cells, which had been stably transfected with the expression vector of C-terminal FLAG-tagged human GHS-R1a or empty vector, respectively, were cultured in Dulbecco's modified Eagle's medium (DMEM) (Wako Pure Chemical Industries) supplemented with 10% fetal bovine serum (FBS) (Invitrogen, Carlsbad, CA, USA) at 37ºC under 5% CO 2 in air. Transfection was performed by using PEI Max (Polysciences, Inc., Warrington, PA).

SIRT1 activity assay
293-GHS-R cells or 293-Mock cells were seeded in 24-well plates and cultured for 24-hour. The media was changed to serum-free DMEM and incubated overnight. After that, the cells were pretreated with rikkunshito for 1 h, and then stimulated with ghrelin for an additional 6 h. SIRT1 activity in the cell lysates was measured using CycLex SIRT1/Sir2 Deacetylase Fluorometric Assay kit (CycLex Co., Ltd).

Ca 2+ flux assay
A Ca 2+ flux assay was performed using HEK293A cells that stably expressed human GHS-R1a and mock cells. The cells were seeded in 96-well plates and treated with ghrelin (1-100 nmol/L), rikkunshito (100 g/mL), or vehicle. The intracellular Ca 2+ was measured with the FLIPR Calcium 5 Assay kit (R8185, Molecular Devices, LLC, Sunnyvale, CA, USA) in accordance with the manufacturer's instructions. The increase in maximal response and the area under the curve (AUC) of Ca 2+ were evaluated.

cAMP assay
293-GHS-R cells were seeded in 24-well plates and cultured for 24-hour. The media was changed to serum-free DMEM and incubated overnight. After that, the cells were pre-treated with 300μM IBMX for 30 min, followed by ghrelin for 30 min, and rikkunshito and SP-A for 90 min in the presence of IBMX. The intracellular cAMP concentrations were measured with a Direct cAMP ELISA kit (Enzo Life Sciences, Farmingdale, NY, USA) in accordance with the manufacturer's instructions.
After washing with phosphate-buffered saline, the cells were treated with Lysis Buffer (AdipoGen International, Inc., San Diego, CA, USA) for 5 min. After processing, the obtained cell lysate was stored at -80°C until measurement. Levels of SIRT1 protein (intracellular, human, AdipoGen International, Inc.) and phosphorylated AMPK (AMPKALPHA PT172, Invitrogen, Life Technologies, Grand Island, NY, USA) were measured using an ELISA.

Impedance-based cell assay
The impedance-based cell assay was performed using the CellKey TM system (Molecular Devices, LLC). The CellKey TM assay system can detect electrical impedance across monolayer cells embedded in electrical fields in each well of 96-well dishes, and these changes indicate changes in intracellular signaling. Ghrelin and rikkunshito were applied to 96-well plates seeded with human GHS-R1a-expressing HEK293A cells, and agonist-induced changes in cellular impedance were measured with the system.

Caspase-3/7 activity assay
GHS-R1a-expressing HEK293A and mock cells were exposed to H 2 O 2 (0.15 mmol/L) for 20 hours. Cell apoptosis was determined with a caspase-3/7 activity assay using IncuCyte (Essen BioScience, Inc, Ann Arbor, MI, USA). The data are expressed as the ratio of fluorescence intensity in caspase-3/7-positive cells treated with 100 nmol/L ghrelin and/or 100 g/mL rikkunshito.

Statistical analysis
Sample size was based on preliminary experiment. Animals were randomly allocated to experimental groups to be no difference in the body weight. Animal studies excluded aging score were not blinded. Values for individual groups are shown as the mean ± standard error (SE). To assess differences among groups, a Student t-test, a multi-group Dunnett test, a Bonferroni test, or Chai-square test for independence was performed. Mortality data were compared with log-rank tests and Gehan-Breslow-Wilcoxon tests. Values of P < 0.05 were considered statistically significant.
Supplementary Figure1. Ghrelin-related factors in klotho-deficient mice. (a) The plasma concentrations of acyl ghrelin, growth hormone (GH), desacyl ghrelin, and corticosterone increased, and the insulin-like growth factor (IGF-1), insulin, glucose, and acyl ghrelin/desacyl ghrelin (A/D) ratio decreased in 5-week-old klotho-deficient mice under fed or/and fasted conditions. These hormonal changes were consistent with those observed in cachexia. (b) The hypothalamic gene expression of neuropeptide Y (NPY) and agouti-related peptide (AgRP) increased, and proopiomelanocortin (POMC) and orexin expression decreased in fasted klotho-deficient mice, suggesting some response to starvation that failed to ameliorate the cachectic condition. CRF: corticotropinreleasing factor. * P < 0.05, ** P < 0.01 (n=9-11). x 3 x 2 x 1.5 x 1/1.5 x 1/2 x 1/3 Supplementary Figure 4. Hypothalamic inflammatory and appetite-related peptide gene expression in klothodeficient mice. (a) Klotho-deficient mice showed increased hypothalamic gene expression of interleukin-6 (IL-6) and tumor necrosis factor-α (TNF-α), which were not affected by rikkunshito (RKT; 1000 mg/kg, p.o.) and atractylodin (ATR; 1 mg/kg, p.o.) administration for 11 days. There were no differences in gene expression of microglia marker, ionized calcium binding adaptor molecule 1 (Iba-1) and peripheral-type benzodiazepine receptor (Tspo) in hypothalamus. * P < 0.05 (n=9-10). (b and c) Hypothalamic appetite-related peptide gene expression levels were measured using a microarray analysis. Data are shown as the ratio of the expression levels in klotho-deficient mice to wild-type mice. Modest increases in arginine vasopressin (AVP) and POMC and a decrease in IGF-1 after treatment with rikkunshito and atractylodin were observed. Twenty-three-week-old SAMP8 (P8) and SAMR1 (R1) mice were given rikkunshito (RKT)-containing chow or control chow. The rates of change in body weight and food efficiency, which were calculated as body weight gain per food intake every five weeks, decreased in SAMP8 mice. These were not affected by RKT treatment. (c and d) There were no differences in anxiety-like behavior in an open-field test (c) or memory disturbance in a step-through passive-avoidance test (d), although the difference between 39-or 40-week-old SAMP8 and SAMR1 mice failed to reach statistical significance. ** P < 0.01 (n=17-20).  Figure 6. Ghrelin-related and inflammatory factors in SAMP8 mice. (a) Twenty-three-week-old SAMP8 (P8) and SAMR1 (R1) mice were given rikkunshito (RKT; 1%)-containing chow or control chow for 19 weeks. The plasma concentrations of growth hormone (GH), desacyl ghrelin, and insulin-like growth factor-binding protein (IGFBP-3) increased, and insulin and leptin decreased in P8 mice compared to R1 mice. RKT treatment increased plasma insulin-like growth factor (IGF-1) concentration, but there were no differences between P8 and RKT-treated SAMP8 mice (P8 + RKT) on the other parameters. A/D: acyl ghrelin/desacyl ghrelin. (b) SAMP8 mice showed increased hypothalamic gene expression of interleukin-1β (IL-1β), tumor necrosis factor-α (TNF-α), and ionized calcium binding adaptor molecule 1 (Iba-1), which were not affected by RKT (1%) administration. * P < 0.05, ** P < 0.01 (n=15-19). (a) An impedance-based cell assay was conducted using the CellKey TM assay system. HEK293A cells stably expressing human GHS-1a receptor were pretreated with RKT (5 and 50 g/mL) or vehicle. Then, ghrelin (1 nmol/L) was applied to cells for 600 s, and electrical impedance, which is induced by a change of intracellular signaling, was detected. (b) GHS-R1a-expressing HEK293A (293-GHS-R) cells and mock (293-Mock) cells were exposed to H 2 O 2 (0.15 mmol/L) and treated with 100 nmol/L ghrelin and/or 100 g/mL RKT for 20 hours. Cell apoptosis was determined with a caspase-3/7 activity assay. Rikkunshito potentiated the cellular response to ghrelin (a) and inhibited oxidative cell death (b). ** P < 0.01 (n=3). (a) Rikkunshito (RKT) upregulated SIRT1 activity in HUVECs, which was blocked by the ghrelin antagonist (D-Lys3)-GHRP-6. * P < 0.05, ** P < 0.01 (n=6). (b and c) Ghrelin, RKT, and atractylodin elicited increases in SIRT1 protein expression in HUVECs. The effect of RKT and atractylodin was inhibited by treatment with (D-Lys3)-GHRP-6 or GHS-R inverse agonist (SP-A). * P < 0.05, ** P < 0.01 (n=6). (d) RKT upregulated phosphorylated AMPK expression in HUVECs, which was blocked by (D-Lys3)-GHRP-6. ** P < 0.01 (n=6). (e) The SIRT1 protein expressions in HUVECs were increased by AMPK activator AICAR and decreased by AMPK inhibitor Compound C. * P < 0.05, ** P < 0.01 (n=6). The levels of SIRT1 protein and phosphorylated AMPK in cell lysate were measured using an enzyme-linked immunosorbent assay in this study.  Figure 13. Ghrelin signaling as a mimetic of caloric restriction (CR). Ghrelin and the ghrelin signaling potentiators rikkunshito and atractylodin increased sirtuin1 (SIRT1) activity through cAMP-CREB pathway or phosphorylated adenosine monophosphate-activated protein kinase (AMPK) in GHS-R expressing cells. Rikkunshito increased hypothalamic SIRT1 activity and ameliorated inflammatory activation of microglia, leading to the improvement on age-related diseases and survival in klotho-deficient mice, SAMP8 mice, and ICR mice, three different animal models on human aging that are characterized by ghrelin resistance. These results indicate that ghrelin secreted from stomach in response to fasting and CR may underlie the beneficial effects of CR on aging through SIRT1 pathways in the hypothalamus. The potentiation of ghrelin signaling may be useful to delay age-related diseases and functional decline, and to extend health-and life-span.

Klotho
The pathology of mice that died or were euthanized at the end of the experimental period in a survival study was observed. Calcification was observed in several tissues, such as the heart, stomach, aorta, and kidney of klotho-deficient mice. Histological score of calcification in heart is shown in Figure 1f. These findings suggest that a major cause of death in klothodeficient mice was systemic calcification. Other pathological changes in the spleen, thymus, and mesenteric lymph node of klotho-deficient mice were also observed. ** P < 0.01 vs. Wild. The observation of pathology after death in the survival study showed the development of tumors in the lung and liver of ICR mice. There was no significant difference in the occurrence of tumors between ICR and rikkunshito (RKT; 1%)-treated ICR mice.